The MPLS WG Archive

["Darek Skalecki" <dareks@nortelnetworks.com>]

MPLS wrote:

> Hi,
>
> This is in reference to implementation of IGP extensions to support TE.
>
> If the IGP advertises the Reserved bandwidth per priority, then what are
> the mechanism/equations to deduce the Unreserved bandwidth per priority?
> Also how about the reverse mapping (i.e from Reserved bandwidth per
> priority how do we et the Unreserved b/w per priority).
>
> Thanks in advance,
> gargi

There are 2 priorities defined in MPLS: holding priority and setup
priority. Setup priority specifies how important it is to establish an LSP,

while holding priority specifies how important it is for an already
established LSP to hold on to its reserved resources. Both priorities have
a range of 0 (highest priority) to 7 (lowest priority). Relationship
between the priorities is the following: an LSP with higher (numerically
lower) setup priority can bump (preempt) an LSP with lower (numerically
higher) holding priority in order to acquire its reserved resources.

Note that holding priority should never be lower (numerically higher) than
setup priority or continuous bumping (preemption) may occur.

 If you know the reserved bandwidth by holding priority (BWH[h], h = 0..7)
then you can compute unreserved bandwidth by setup priority (BWS[s], s =
0..7) using the following formula:

BWS[s] = MRB - SUM_h=0_to_s(BWH[h])

where MRB is the maximum reservable bandwidth, and SUM_h=0_to_s is a crude
representation of mathermatical summation notation.

For example,

BWS[3] = MRB - (BWH[0] + BWH[1] + BWH[2] + BWH[3])
BWS[0] = MRB - (BWH[0])

If people agree then maybe the formula should make its way into one of the
drafts (OSPF/ISIS TE extensions or CR-LDP)?

Darek